Using the results from the previous question (11) re-run the query and include a count of event filtered to action that matches add to cart, starting from 1 month ago ending now. Use a table view, or interact with the Sankey to determine which referrer, device type and region combination have the most add to cart events. You can do this by toggling to a table view, or you can interact directly with the Sankey visualization by clicking on a node and selecting "filter to".
When looking at the results of a table view or a Sankey, you’ll notice that Scuba automatically puts the results with the highest value at the top. That way if you’re interacting with the Sankey to “filter to” by clicking you can easily construct a query that narrows down on the users or events that you want to dig into. This also works with table view.
Note: The numbers and result does not need to be an exact match as the data changes over time. The query construction is what is most important. See images below.
*Click to enlarge*
If you ran the query as a table view you can easily see that the top result which is Other Web, Android and US & Canada in this case.
If you were to interact with the Sankey by hovering over each node and determining which had the highest count then clicking and selecting “filter to” you will notice the measure looks different because it incorporates the “filter to” criteria in the measure.